Status Monitoring Apparatus, Status Monitoring Method, and Program

1. A status monitoring apparatus comprising: an information obtaining module that obtains detected information items that are continuously detected; an image data generating module that partitions the detected information items, which are obtained by the information obtaining module, by frames at predetermined time intervals, and configured to generate, for each frame, two-dimensional image data representing a distribution of the detected information items in the frame, with an axis of abscissa being a time axis and an axis of ordinate being an axis showing a value or an amount of the detected information items; and a status detecting module that performs a machine learning based on the two-dimensional image data which are sequentially generated by the image data generating module, so as to generate a correct label for judging whether the two-dimensional image data show a failure or a warning sign of failure, and configured to judge that a failure or a warning sign of failure has occurred in the detected information items, when the two-dimensional image data corresponding to the correct label are generated.

2. The status monitoring apparatus according to claim 1 , wherein the image data generating module generates, as the two-dimensional image data, image data representing a polygonal line graph sequentially connecting, by line segments, the detected information items adjacent to each other in the frame.

3. The status monitoring apparatus according to claim 1 , wherein the image data generating module partitions the detected information items by two frames continuously adjacent to each other, sequentially shifts a first frame of the two frames at detection intervals of the detected information items until the first frame is positioned to overlap with a next frame, extracts maximum values and minimum values of the detected information items at respective positions before the frame is shifted and respective positions after the frame is shifted, chronologically arranges the extracted maximum values and the minimum values in an ascending order of the detection timings, so as to generate image data showing areas like strips such that an area between a boundary line connecting the maximum values along a polygonal curve and a boundary line connecting the minimum values along a polygonal curve differs from other areas, and provides the image data as the two-dimensional image data corresponding to the first frame.

4. The status monitoring apparatus according to claim 1 , wherein the image data generating module partitions the detected information items by two frames continuously adjacent to each other, extracts, in the respective frames, maximum values and minimum values of the detected information items, sets the maximum value and the minimum value of the detected information items in a first frame of the two frames, at a front end of the axis of abscissa of the first frame, sets the maximum value and the minimum value of the detected information items in a next frame, at a rear end of the axis of abscissa of the first frame, generates image data showing areas like strips such that an area between a boundary line connecting the maximum values by a line segment and a boundary line connecting the minimum values by a line segment differs from other areas, and provides the image data as the two-dimensional image data corresponding to the first frame.

5. The status monitoring apparatus according to claim 1 , wherein the detected information items are current values of a current supplied to a compressor of a refrigeration circuit constituting an air conditioner, pressure values of a refrigerant circulating in the refrigeration circuit, acceleration values of an expansion valve of the refrigeration circuit, current values of a current supplied to a blower constituting the air conditioner, or current values of a current supplied to a pump for conveying a liquid.

6. A status monitoring method comprising: an information obtaining step which obtains detected information items that are continuously detected; an image data generating step which partitions the detected information items, which are obtained in the information obtaining step, by frames at predetermined time intervals, and generates, for each frame, two-dimensional image data representing a distribution of the detected information items in the frame, with an axis of abscissa being a time axis and an axis of ordinate being an axis showing a value or an amount of the detected information items; and a status detecting step which performs a machine learning based on the two-dimensional image data which are sequentially generated in the image data generating step, so as to generate a correct label for judging whether the two-dimensional image data show a failure or a warning sign of failure, and judges that a failure or a warning sign of failure has occurred in the detected information items, when the two-dimensional image data corresponding to the correct label are generated.

7. A program executable in a computer to perform: an information obtaining step which obtains detected information items that are continuously detected; an image data generating step which partitions the detected information items, which are obtained in the information obtaining step, by frames at predetermined time intervals, and generates, for each frame, two-dimensional image data representing a distribution of the detected information items in the frame, with an axis of abscissa being a time axis and an axis of ordinate being an axis showing a value or an amount of the detected information items; and a status detecting step which performs a machine learning based on the two-dimensional image data which are sequentially generated in the image data generating step, so as to generate a correct label for judging whether the two-dimensional image data show a failure or a warning sign of failure, and judges that a failure or a warning sign of failure has occurred in the detected information items, when the two-dimensional image data corresponding to the correct label are generated.